I've been doing PostgreSQL full time since 2013. I do consulting, training, and auditing out of Paris, France. I am also the co-founder and co-organizer of pgDay Paris.
PostgreSQL offers different kinds of buffers used for caching and most of them are configurable. But how should they be configured? How big should my shared_buffers be? Do I need to do anything special if I use a lot of temporary tables? Will a sequential scan on a huge table ruin my cache? All these questions will be answered, and we will also talk about considerations for configuring checkpoints and the background writer and how that affects the choice of buffer size.